lable at ScienceDirect

Journal of South American Earth Sciences 30 (2010) 176e188

Contents lists avai

Journal of South American Earth Sciences

journal homepage: www.elsevier .com/locate/ jsames

Geochemical behavior and dissolved species control in acid sand pit lakes,Sepetiba sedimentary basin, Rio de Janeiro, SE e Brazil

Eduardo D. Marques a,b, Sílvia M. Sella a, Edison D. Bidone a, Emmanoel V. Silva-Filho a,*

a Instituto de Química, Universidade Federal Fluminense, 24020-141 Centro, Niterói, BrazilbGeological Survey of Brazil, CPRM, 30140-002 Belo Horizonte, Brazil

a r t i c l e i n f o

Article history:Received 22 July 2009Accepted 13 April 2010

Keywords:HydrogeochemistryPluviositySand pit lakesAcidificationSepetiba basin

* Corresponding author.E-mail address: geoemma@vm.uff.br (E.V. Silva-Fi

0895-9811/$ e see front matter � 2010 Published bydoi:10.1016/j.jsames.2010.04.003

a b s t r a c t

This work shows the influence of pluvial waters on dissolved components and mineral equilibrium offour sand pit lakes, located in the Sepetiba sedimentary basin, SE Brazil. The sand mining activitiespromote sediment oxidation, lowering pH and increasing SO4 contents. The relatively high acidity ofthese waters, similar to ore pit lakes environment and associated acid mine drainage, increasesweathering rate, especially of silicate minerals, which produces high Al concentrations, the limitingfactor for fish aquaculture. During the dry season, basic cations (Ca, Mg, K and Na), SiO2 and Al show theirhigher values due to evapoconcentration and pH are buffered. In the beginning of the wet season, thedilution factor by rainwater increases SO4 and decreases pH values. The aluminum monomeric forms (Al(OH)2þ and Al(OH)2þ), the most toxic species for aquatic organisms, occur during the dry season, whileAlSO4

þ species predominate during the wet season. Gibbsite, allophane, alunite and jurbanite are thereactive mineral phases indicated by PHREEQC modeling. During the dry season, hydroxialuminosilicateallophane is the main phase in equilibrium with the solution, while the sulphate salts alunite andjurbanite predominate in the rainy season due to the increasing of SO4 values. Gibbsite is also in equi-librium with sand pit lakes waters, pointing out that hydrolysis reaction is a constant process in thesystem. Comparing to SiO2, sulphate is the main Al retriever in the pit waters because the most samples(alunite and jurbanite) are in equilibrium with the solution in both seasons. This Al hydrochemicalcontrol allied to some precaution, like pH correction and fertilization of these waters, allows theconditions for fishpond culture. Equilibrium of the majority samples with kaolinite (Ca, Mg, Na diagrams)and primary minerals (K diagram) points to moderate weathering rate in sand pit sediments, whichcannot be considered for the whole basin due to the anomalous acidification of the studied waters.

� 2010 Published by Elsevier Ltd.

1. Introduction

The acidification of natural waters is of great concern to themining industry around the world. When sulphide minerals areexposed to the atmosphere, in the presence of water, sulphideoxidation may occur, producing sulphuric acid and releasing tracemetals and other pollutants to the water (Modis et al., 1998;Bachmann et al., 2001; Ramstedt et al., 2003; Denimal et al., 2005;Pellicori et al., 2005; Sheoran and Sheoran, 2006). Due to itscomplexity, many researchers have investigated the mechanisms ofmetals leaching in natural waters (Alpers and Blowes, 1994; Jamborand Blowes, 1994; Plumlee and Logsdon, 1998; Ulrich et al., 2006).

The Sepetiba basin, located in Rio de Janeiro western metro-politan region, has a great potential for sand mining. At the end of

lho).

Elsevier Ltd.

the 1960s, with increasing civil construction Seropédica-ItaguaíSandMining District became the main sand supplier for that region(about 70%), with more than 80 active sand pit mines. The pit lakeshave a total area of about 40 km2 (Fig. 1), average depth of about28m and the total volume (the entire SandMining District) is about540 km3.

The sand extraction process is performed by the removal ofsurface sedimentary layers, exposing Piranema aquifer phreaticsurface, which fills up the pit holes. However, these activitiesgenerate impact on thewater quality due towater table drawdown,which may suffer contamination by fuel oil from dredge machineand or domestic/industrial wastes (Berbert, 2003; Marques et al.,2008).

The sand pit lakes show peculiar low pH values and this acidi-fication process is similar to the one from the sulphide ore pit lakes,that is, it is given by the sulphide phase oxidation. The geologicalenvironment of the sand pit lakes is the key for water acidification.It is a sedimentary basin, dating rom the Tertiary/Quaternary (Góes,

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188 177

1994). The geological history of this region includes ancientmangrove and swamp environments (Berbert, 2003), which givethe conditions for organic matter accumulation and sulphidespecies formation (sulphidric gas and neo-formed pyrite), thatwhen exposed by the sand extraction process may undergooxidation, decreasing the pH values of sand pit lakes waters.

Then, the pit lakes generated by ore mining as well as by sandmining may become social amenities or environmental nuisancesdepending on lake water chemistry. Many recent mine manage-ment strategies favor the progressive rehabilitation of pit lakes intopublic amenities that will exist long after mine closure, as recrea-tional areas, theme parks, scientific research facilities, and wastedisposal sites (Castendyk and Webster-Brown, 2007; Stottmeisteret al., 1999; Fisher and Lawrence, 2000; Kennedy-Perkins, 2002).

Fig. 1. The Piranema aquifer boundaries and studied area location. The four sa

In order to create social and/or economic amenities and avoidenvironmental problems, studies in sand pit lakes are in progress togive a useful destination to these artificial lakes, and one of thepossibilities is fishpond culture. This possibility could be applied insand pit lakes, which, as in ore mining, are left behind when thisactivity is finished.

Observing the composition of Sepetiba basin sediments, whichhave quartz and feldspar as main minerals (almost 96% of the basinsediments), it is clear that groundwater composition is also poor,that is, only the major compounds are found in those waters (Na, K,Ca, Mg and Cl, Barbosa, 2005). However, with water acidification,new compounds are likely to become dissolved, especially Al andSO4 in this area. The region pluviosity showed to be an importantcontrol factor of the dissolved contents of sand pit lakes, mainly

nd pit lakes studied are highlighted by thicker lines on the lower figure.

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188178

of Al, whose species may limit the use of these pit lakes asfishpond culture.

The aim of this study is to characterize sand pit lakes compo-sition and its sources, the factors (including rainwater seasonality)and geochemical processes (including mineral equilibriumapproach) that control its composition and the behavior of themain dissolved species. Special emphasis is given for Al availability,which has an essential role in the introduction of the fishaquaculture.

Fig. 3. Average values of evapotranspiration and temperature from Sepetiba region(modified from Carvalho et al., 2006) during the sampling period.

2. Study area

The Sepetiba Sedimentary Basin occupies an area of about 4% ofthe Rio de Janeiro State, and its main tributary is the Guandu River,which is originated in the Serra do Mar mountains. The Guandudrainage basin occupies an area of 2000 km2, 90% of which hasfeatures of alluvial plain deposits (SEMA, 1996). The studied area(Fig. 1) lies within this plain and is located at the UTM coordinatesNorth 7,470,000; 7,478,000 and UTM coordinates East 630,000;638,000. The Guandu River receives water from Paraíba do Sul riverdiversion, and flows to the Guandu Water Treatment Station, thelargest of Latin America (Rios and Berger, 2002).

There are two well-known precipitation periods based on thehistorical monthly averages between 1977 and 2005 (Santa CruzPluviometric Station) and Fig. 2 shows the pluviometric precipita-tion during all the sampling campaigns. The pluviometric precipi-tation is higher from October to March and lower from April toSeptember. An interesting feature of Sepetiba region is its highevapotranspiration, with annual average (from 1961 to 1985) of3.6 mm day�1 (Carvalho et al., 2006; Fig. 3).

The geology of the studied region is composed of Tertiary/Quaternary sediments from alluvial environment (fluvial, fluvial-lacustrine and fluvial-marine) deposited on Precambrianbasement. These sediments form the Piranema Formation (Góes,1994) and are represented by two units. The lower unit presentsPleistocene sandy facies, with medium to coarse texture andgenerally basal gravel, and mineralogy essentially quartz-feld-spathic. The upper unit, also called alluvial cover, is composed ofHolocene silt-clay facies. Sediment cores carried out at the regionpointed out mean thickness ranging about 35 and 40 m, reachingdepth larger than 70 m in some cases.

The mineralogy of these sandy sediments was characterized byBerbert (2003), who reported 82% quartz, 14% feldspars (about 80%of K-feldspars and 20% plagioclase) and 2% micas and rock frag-ments, classifying the sandy fraction as sub-arcosian.

Fig. 2. Rainfall from Sepetiba region (Santa Cruz plu

The Sepetiba sedimentary basin also has some features, like highporosity and good permeability, which give the conditions forwater accumulation and transmission, characterizing the PiranemaFormation as an aquifer called Piranema Sedimentary Aquifer(Tubbs, 1999). This sedimentary aquifer system has an area of about350 km2 (70% of the area shown in Fig. 1) and is located approxi-mately 60 km west from Rio de Janeiro city. The free aquifersystems recharge is distributed upon its occurrence area, trendingto highest potenciometric level as high as the regional topography.So, the flux direction is controlled by topographic irregularity. Thewater table level ranges from 3 to 7.5 m, depending on the weatherseason.

The soil covers, originated from crystalline rocks, could generatean aquifer system with similar characteristics of a sedimentaryporous aquifer (colluvial deposits) and gradually, as depthincreases, passing into fractured systems. Together, the fracturedaquifers and soil covers are responsible for 30% (150 km2) of thearea (Fig. 1). Intercommunication among sedimentary, fracturedand colluvial aquifers could increase the regional groundwaterpotential and determines the aquifer recharge and flow patterns(ELETROBOLT, 2003).

2.1. Strategic importance of the Piranema aquifer

The Guandu hydrographic basin has strong dependence on thewater diversion from Paraíba do Sul River. While the transpositionflow average is of about 166 m3 s�1, the contribution of its ownbasin is of about 3.18 m3 s�1. With the absence of other significant

viometric station) during the sampling period.

Table 1Average; standard deviation; maximum and minimum concentration (mg L�1); pH and electrical conductivity (EC) from studied sand pit water compared to a groundwaterstudy in the same region.

Ca Mg Na K Total Fe Mn Al SiO2 SO4 Cl pH EC (mS cm�1)

Sand Pit Lake 1Average 6.9 2.7 27.1 3.1 0.2 0.4 2.5 26.3 61.5 22.1 3.8 295SD 5.4 2 14 1 0.1 0.1 5.1 9.6 19 8 0.3 34.3Maximum 20.95 5.6 66.08 5.17 0.46 0.6 14.7 34.4 95.95 33.25 4.93 329Minimum 1.3 0.16 6.35 1.5 0.02 0.06 0.02 2.6 1.34 9.88 3.11 208Sand Pit Lake 2Average 2.9 1.5 26.1 2.7 0.3 0.1 2.8 26.6 29.8 26.5 4.4 194SD 1.2 1 7.5 0.5 0.7 0.01 5 4.6 8.6 4.6 0.3 30.5Maximum 4.56 3.12 63.33 4.65 2.05 0.22 13.7 32.8 41.05 33.09 5.2 223Minimum 1.03 0.07 10 1.86 0.004 0.1 0.01 16.36 0.91 17.7 3.96 121Sand Pit Lake 3Average 1.9 0.5 16.8 2 0.08 0.13 1.5 21.1 3.3 25.6 4.5 127SD 3.2 0.4 6.4 0.6 0.11 0.03 4.1 5.8 0.9 4.2 0.2 8.8Maximum 10.91 1.99 64.13 3.09 0.3 0.18 12.3 29.9 4.77 30.53 5.08 143Minimum 0.27 0.01 0.02 0.38 0.001 0.05 0.01 9.3 0.11 18.27 4.14 111Sand Pit Lake 4Average 5.7 3.6 33.3 3.9 0.1 0.4 2.9 25.5 60.9 36.3 4.5 301SD 2.4 2.3 10.8 0.7 0.1 0.1 5.7 5.3 23.8 12 0.5 69.9Maximum 9.83 6.49 71.33 5.83 0.66 0.65 14.41 33.8 97.64 58.44 5.19 388Minimum 2.48 0.3 16.4 2.5 0.02 0.24 0.04 14 1.82 16.9 3.68 207Groundwater data from Barbosa (2005)Average 3.6 3.6 31.1 4.8 0.8 0.2 <0.015 n.a. 6.5 43 5.0 203.0SD 2.3 2.5 9.5 2.3 2.9 0.2 e e 8.6 14.6 0.3 10.0Maximum 9.96 11.98 56 11.46 14.3 0.8 e e 39.3 76.4 5.66 230.0Minimum 1.1 1.1 18.4 0.5 0.003 0.1 e e 0.8 26.6 4.9 20.0

n.a. ¼ non-analyzed

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188 179

water body in that region, the Paraíba do Sul River is practically theonly water supply source to industrial and domestic use in Rio deJaneiro Metropolitan Region (Ottoni et al., 2002; Rios and Berger,2002).

The Piranema Aquifer, which presents an area of about 500 km2

and an outflow available about 1.6 m3 s�1, is the main watersupplier to the Seropédica-Itaguaí district. Although presentinglimited water availability, compared to Guandu river water trans-position volume, this sedimentary aquifer has strategic importancefor Rio de Janeiro metropolitan region. Its water reservoir is theonly alternative that can be used during scarce period of potablewater or environmental accidents, when the conventional systemsof water supply are endangered.

3. Materials and methods

Water samples were collected (4 m below the water surface)from November 2003 to November 2005 in four sand pit lakes(Fig. 1), using 2 L Van Dorn bottle and stored in clean polyethylenebottles. In general, 56 water samples were collected bimonthly in14 sampling campaigns. Electrical conductivity, temperature andpH were measured using WTW-LF electrodes (model 330) at thesampling sites.

In the laboratory, the samples were filtered through 0.45 and0.22 mm filters (MILLEX�-GS MILLIPORE filters). This procedure wasnecessary in order to assure clay and colloids fraction separationfrom the samples to be analyzed. The filtered samples were dividedinto three equal aliquots of 50 mL. The first aliquot (0.45 mm) wasused for silica analysis, the second one (0.22 mm) was acidified (pH

Table 2Average ions concentration (mg L�1); pH; temperature and pluviometric precipitation (P

Ca Mg Na K Total Fe Mn Al

Wet season 3.32 2.06 31.48 3.40 0.15 0.31 0.17Dry season 5.84 3.12 39.19 4.48 0.16 0.31 3.55

PPT ¼ pluviometric precipitation.

1) and used for metal analysis, and the third one (0.22 mm) wasused to determine anions and silica content.

Chloride and sulphate determinations were carried out byionic chromatography (SHIMADZU LC-10AD, conductivity detectorCDD-6A). The determination of Ca2þ, Mg2þ, Kþ and Naþ wasperformed by flame atomic absorption spectrometry (VARIANSPECTRAA-300). Aluminum, total iron and manganese determina-tions were carried out by Inductively Coupled Plasma OpticalEmission Spectrometry (JOBIN YVON e HORIBA, ULTIMA 2), whosedetection limits were 0.2, 0.2 and 0.05 mg L�1, respectively. Silicawas determined by specific spectrophotometric methods(Grasshoff et al., 1983) and analyzed with a HITACHI, U-1100spectrophotometer. For mineral equilibrium and dissolved speciescalculations, the PHREEQC modeling program (Parkhurst, 1995)was used WATEQ4F mineral data base (Ball et al., 1987) ascomplement.

4. Results

Table 1 shows the average (n ¼ 14), maximum and minimumvalues and standard deviation of dissolved Ca, Mg, Na, K, total Fe (IIand III),Mn, Al, SiO2, Cl, SO4 concentrations (mg L�1); pH and electricconductivity (mS cm�1) in the sand pit waters. The groundwaterfrom wells in the studied region (Barbosa, 2005) shows higher pHvalues compared to pit lake waters. Sand pit lake 1 showed thelowestmeanpH value (3.8), while sand pit lakes 3 and 4 showed thehighest ones (4.5). The electric conductivity values ranged from 111(sand pit 3) to 388 mS cm�1 (sand pit lake 4), suggesting lowmineralization of these waters (Marques et al., 2004).

PT) at the dry and the wet season.

SiO2 SO4 pH Cl EC (mS cm�1) T(�C) PPT (mm)

23.29 35.27 4.28 30.05 240.45 29.35 141.8325.32 18.58 4.61 30.31 225.03 24.68 69.37

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188180

In comparison to a study made by Barbosa (2005), the chemicalcomposition of the sand pit lakes showed slightly lower concen-trations of Ca, Mg, Na, K, dissolved Fe and Cl and about one order ofmagnitude higher for Al and SO4 (Table 1). In general, among thefour sand pit lakes, the highest average values of the studiedparameters were found in sand pit lake 4 and the lowest ones insand pit lake 3, the most recently opened, which also showed thelowest SO4 content. Highest contents of Al, total Fe and SiO2 wereobserved in the dry season, as opposed to the SO4 and Cl concen-trations, which showed their highest values at the wet season. Dueto the low pH found in the sand pit lakes, the absence of bicar-bonate and carbonate ions was assumed, indicating that thosewaters may be classified as NaeSO4eCl facies type, reflecting itsmain components (Marques et al., 2008).

Therefore, the anomalous contents of Al (14.7; 13.7; 12.3 and14.4 mg l�1 in sand pit lakes 1, 2, 3 and 4, respectively), togetherwith the high SO4 content and the fact that SiO2 is mainly associ-ated to colloidal fraction, due to the pH range of these waters, arethe outstanding features of the sand pit lakes.

5. Discussion

5.1. The water chemistry control of sand pit lakes

As seen in Item 2, the Sepetiba basin has two well-knownprecipitation periods. So, it is expected that in the wet season(OctobereMarch), the dilution factor by rainwater occurs, while theopposite phenomenon, the evapoconcentration, has effect in the

Fig. 4. pH and electrical conductivity EC be

dry season (AprileSeptember). Evapoconcentration is an importantprocess that affects the chemical composition of lakes located inhydrologically closed basins, where the primary route for water lossis evaporation (Eary, 1999). The evapotranspiration in the dryseason can also reach high values (>6 mm.day�1, Carvalho et al.,2006) in Sepetiba basin, and the studied sand pit lakes are under-going this process for more than 10 years, with exception of themost recent opened sand pit lake (Santa Helena, 2003). Table 2shows average ions concentration in the wet and the dry seasonsfor sand pit lakes studied. The majority of the analyzed ions hadsignificant variation in their content, showing the importance ofrainwater seasonality in the chemical composition in these pitlakes. The exceptions were Fe, Mn and Cl contents, which hadalmost no variation during the pluviometric regime.

Therefore, the chemical composition of sand pit lakes water iscontrolled by region pluviosity and mining activities. Together,these factors control physicalechemical parameters variations,especially pH and electrical conductivity (EC). The sand pit lakes EChas relatively low values (groundwater in the studied area andother regions of Rio de Janeiro have values above 200 mS cm�1e seeTable 1 and Bidone et al., 1999), which practically does not varybetween seasons. The reason for the low EC values is the miner-alogy, mostly quartz and K-feldspar, which are weathering resis-tant. On the other hand, the reason for the low pH values is that theweathering reactions rate is not enough to consume hydrogen ions.Fig. 4 shows pH and EC values along sampling campaigns. It canbe seen that pH values are different among pit lakes until December2004, when they became similar. As for EC, an abrupt fall in its

havior along the sampling campaigns.

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188 181

values for sand pit lakes 1, 2 and 4 could be observed in August2004, coinciding with low rainfall event in the same period (Fig. 2).Fig. 5 shows the trend of EC in the pH range studied at the dry andthe wet season. It can be seen that EC shows high values in bothseasons and pH for wet season presents the lowest values and thehighest ones for dry season. The pH behavior could be explained bythe ions concentration in the water, represented by EC.

As mentioned before the sand extraction process, which doesnot stop its activities even during the rainy season, leads to low pHvalues, contributingwith primaryminerals dissolution (weatheringprocess), besides desorption of some compounds in organic matterand clay minerals. Figs. 6 and 7 present the water componentsconcentrations versus EC and pH, in order to discuss the watercompounds behavior. The water compounds Ca, Mg, K and Mnpresent good correlation with EC in the dry season. However,during wet season the prominent compounds are SO4, Al and Mn.Manganese shows low concentration in both seasons and it couldbe related to its great mobility in oxidant environment (Brockamp,1976; Bendell-Young et al., 1989; Bendell-Young and Harvey, 1992).Regarding pH, it is noticed that Ca, K, Na, Mg and Mn have similarconfigurations, that is, high values in both seasons. Silica, which isfound in colloidal state (Item 4), seems to have no direct influenceof pH and shows its high concentrations in the dry season. Despitethe relative high correlation of Al with EC during wet season, itshigher values are shown in the dry season.

The same behavior is shown by Fe, but there is a weak corre-lationwith EC during wet season. Al and Fe, which are concentratedin the water during dry season, have their behavior and sourcelinked to the sand pit lakes acidification processes (Al from sedi-ments weathering, Fe from pyrite in reduced sediments and bothfrom biotite). Considering the same source of Al and Fe, the SO4content reaches its higher values at the wet season (the higher onereaches 95.5 mg l�1) and also exerts important influence in ECduring this period. These results agree with the conclusions of Gray(1996) and Robles-Arenas et al. (2006), who showed that sulphateconcentrations and EC are more consistent indicators of mininginfluence than heavy-metal contents or pH values because they areaffected by environmental fluctuations. It is worth to notice that, aswell as Fe, the same Mn values in the dry and the wet season couldbe explained by the sand extraction process on the reduced

Fig. 5. Electrical conductivity (EC) be

sediments. Considering the non-interruption of sand extractionprocess and Eh values (0.5 V; Marques, unpublished data) in bothseasons, Fe and Mn contents in those sediments (Fe in pyrite andMn adsorbed on the clay minerals and into their structures; Larsenand Mann, 2005) could present constant values in those waters,even with dilution during wet season.

Besides showing some high values, chloride has the samebehavior in both seasons, suggesting its conservative way whichcorroborates the hypothesis of sea-salt spray provenience(Seinfeld, 1986; de Mello, 2001; Silva-Filho et al., 2009), asa consequence of the rainfall direct recharge. However, the Sepe-tiba basin has indications of mangrove environment (ancient coastlines) in its sediments records, supporting the hypothesis that Cl aswell as Na (this one has large values for only sea-salt spray andweathering as sources), Ca andMg (carbonates) could have anothersource in the sediments. Moreover, the low percentage of plagio-clase comparing to alkaline feldspar in the sandy fraction asshown in Item 2 corroborates also the hypothesis of Na sea-saltprovenance.

The concentrations of silica in sand pit lakes are higher thanthose observed in other waters in the same region. In the range ofpH values measured, the concentrations of silica are supersaturatedwith respect to equilibrium with quartz (H4SiO4 activity greaterthan 10�4) and undersaturated with respect to amorphous silica(H4SiO4 activity greater than 2 � 10�3). Therefore, it is possible thatamorphous silica is solubilizing and releasing H4SiO4 to the solu-tion, aside from the possibility that silica is present in colloidalform, as shown by Marques et al. (2008).

In summary, during the dry season, without rainwater dilutionand the influence of evapotranspiration, pH values become higherdue to buffering caused by the large content of dissolvedcompounds in sand pit lakes, namely Ca, Mg, K, Mn, Al and SiO2,(Hem, 1985; Stumm and Morgan, 1996; Miretzky et al., 2001), aswell as SO4, leading to the ion shield phenomenon (Deutsch, 1997).On the other hand, during the wet season the rainwater inputdilutes sand pit water, decreasing major cations (Ca, Mg, K and Na)contents (as seen in Table 2), and increasing hydrogen ions. Thereaction rate of reduced sediments oxidation and acidification isobviously higher than weathering and dissolution processes(Stumm and Morgan, 1996) and this fact could explain the increase

havior in the pH range studied.

Fig. 6. Mg, Na, K, Ca and SiO2 concentration in the EC and pH range along wet and dry season.

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188182

of SO4 and hydrogen ions content during the wet season,comparing to basic ions, and considering the non-interruption inthe mining activity in both seasons.

5.2. The chemical control of Al in the sand pit lakes

The high aluminum concentration in the sand pit lakes ishighlighted by the fact of its potential toxicity. Aluminum in acidhabitats has shown to be toxic to fish, amphibians and

phytoplankton (Driscoll and Schecher, 1990; Birge, 1978; Poleo,1995) and is generally more toxic over the pH range of 4.4e5.4,with a maximum toxicity occurring around pH 5.0e5.2 (Campbellet al., 1983; Klöppel et al., 1997). In freshwater systems aluminumspeciation and solubility are highly pH dependent. Solubility islowest between pH 6 and 7, with 90% of the aluminum existing asa colloidal solid. The solubility increases 100 fold between pH 6and 4.7. The toxicity of Al to fish is primarily due to effects onosmoregulation by action on the gill surface (McDonald et al.,

Fig. 7. Al, Fe, Mn, Cl and SO4 concentration in the EC and pH range along wet and dry season.

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188 183

1989). Al is readily accumulated on and in the gill, but little isfound in blood or internal organs. Thus the embryo is the life stageleast sensitive to Al, whereas the fry stage (small larval fish) is themost sensitive; then sensitivity decreases with age (ASTM, 1992).Aluminum toxicity is interactive with that of the hydrogen ion andusually occurs at pH values ranging from 0.3 to 0.6 pH units abovewhich the hydrogen ion causes some lethality. The toxicity of

aqueous Al is reduced by Ca and dissolved oxygen (Gensemer andPlayle, 1999).

The relative contribution of low pH and elevated Al contents isdifficult to determine and varies between geographic regions(Nordstrom and Ball, 1986). CCREM (1999) points out that the limitof Al to aquatic life is about 0.1 mg L�1, in pH values <6.5, whichcould be an obstacle to the introduction of aquaculture (fish and

Table 3Concentration of Al species in sand pit lakes at the dry and the wet season.

Species (mg l�1) Wet season (%) Dry season (%)

Al3þ 0.118 68.6 2.175 61.2Al(OH)2þ 0.023 13.37 0.810 22.8Al(OH)2þ 0.005 2.91 0.333 9.37Al(OH)30 1.7 � 10�5 e 0.008 0.23AlSO4

þ 0.025 14.53 0.222 6.25Al(SO4)2� 0.001 0.58 0.005 0.14Total 0.172 100 3.553 100

Fig. 8. Saturation index (SI) of the reactive mineral phases indicated by PHREEQCmodel as a function of Al concentrations.

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188184

others), as suggested for the sand pit lakes at the end of sandmining.

We performed a study involving Al speciation that was assessedby PHREEQC model, which calculates the dissolved speciesaccording to Al and SO4 (the last one is the most abundant ion insand pit lakes) content and physicalechemical parameters liketemperature, pH and Eh, in the aqueous system. The resultsobtained are shown in Table 3 where it can be seen that thealuminum species have higher concentrations in the dry season,while Al3þ is the predominant species in both seasons, followed byAl(OH)2þ and Al(OH)2þ species. These monomeric species are morereactive at cell membrane surface of aquatic organisms than poly-meric forms and organically bound Al (Baird, 1998; Gensemer andPlayle, 1999; Camilleri et al., 2003).

In the dry season, the AlSO4þ and Al(SO4)2� species are more

abundant even though the SO4 content in this season is lowercompared to the rainy period, suggesting that this component playsan important role in the aluminum availability of these waters andmay be responsible for the blue color, together with the low sus-pended particulate material observed in the water. Moreover,according to Birchall et al. (1989), Exley et al. (1991), Duan andGregory (1998) and Camilleri et al. (2003), the silica watercontent could control aluminum solubility due to the formation ofhydroxialuminosilicates. The formation of these aqueous species,besides the precipitation of SO4 and silica phases, may help todefine the use of these sand pit lakes for fishpond culture.

In order to give more detailed information about thegeochemical Al control in these waters, we used the PHREEQCmodeling on the data, complemented by WATEQ4F mineral database, which shows the results of saturation index (SI) in thewet andthe dry seasons, and is a convenient parameter for evaluating theproximity to equilibrium in various solubility reactions for aqueoussystems (Eary, 1998). Mineral equilibrium calculations for a watersample are useful to predict the presence of reactive minerals inaqueous systems and estimating mineral reactivity (Deutsch,1997).Ranges of SI ¼ 0 � 0.5 were considered as equilibrium condition ofthe solutionwith respect to a mineral due to inherent uncertaintiesin the calculations of SI (such as calculating ions activities) asdefined by Deutsch (1997) in agreement with other authors.

The equilibrium calculations showed four reactive minerals(minerals precipitated directly from the solution) to be consideredin the system: gibbsite (Al(OH)3), allophane (hydrox-ialuminosilicate e [Al(OH)3](1 � x)[SiO2]x) and sulphate salts alunite(KAl3(SO4)2(OH)6) and jurbanite (AlOHSO4). Fig. 8 shows SI valuesof the consideredminerals in function of Al content in both seasons.The four minerals show an equilibrium zone in the wet season.However, alunite is the most important phase to retrieve Al fromthe solution in this season. Gibbsite, jurbanite and also alunite areoversaturated in some samples (SI > 0 � 0.5), likely indicating thelater precipitation of these phases, that is, theywould limit solutionconcentration of its constituents (Al and SO4) to values that wouldproduce an SI close to zero. Allophane seems to be non-reactive inthe wet season (SI < 0 � 0.5). Nonetheless, this hydrox-ialuminosilicate presents the majority of its plots in the equilibrium

zone during the dry season, coinciding with the increase of SiO2and Al content in this period and gibbsite confirms that hydrolysisreaction is a constant process in the system. Even considering thatSO4 concentration decreases in this period, jurbanite still showsequilibrium with the solution and so does alunite in the over-saturated zone, stressing the important SO4 dynamics in the pitlakes waters. These data point out that SO4 phases are better Alretrievers than SiO2 ones because they take place in the equilibriumzone in both seasons. Moreover the SO4 phases display more plotsin the oversaturated zone, indicating later precipitation, while SiO2phase reaches equilibrium only in the dry season and the majorityof its plots is in the undersaturated zone in both periods.

5.3. An overview of aluminosilicates weathering in sand pit lakessediments by mineral stability diagrams

The simplest process that might regulate trace elementsconcentration in aqueous solution is the equilibrium with respectto a solid phase containing the element as a major component, forexample, dissolved Al concentration and its equilibrium withkaolinite. At H4SiO4 activities above 10�4.2, kaolinite is more stablethan gibbsite, and the opposite is true at H4SiO4 activities below

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188 185

this value (Drever, 1982). In this study, H4SiO4 activities values areabove 10�4.2.

Silicate minerals are chemically complex, commonly containingfour or more elements. Therefore their reactions in aqueous solu-tions are directly dependent upon the concentrations of theseelements in solution and often indirectly dependent upon manyother dissolved species (Fleet, 1984). In the case of sand pit lakes,the intensive hydrolysis reactions take place at the aqueous systemdue to high acidity, which promotes high weathering rates of silicamineral phases. For that reason, the sand pit lakes weatheringmust be unusual compared to other parts of the basin, that is, thesurrounding aquifer.

Fig. 9. Sand pit lakes sampling plots (at the wet and the dry season) in the mineral stabilit1 atm. Dashed lines represent the amorphous silica saturation field.

Mineral stability diagrams (Fig. 9) based on the incongruentsolution of aluminosilicate minerals (mainly feldspars and micas,up to 12% of sandy fraction) was assembled, in order to assess themineral weathering rate in sand pit lakes. The reactions on whichthe diagrams are based contribute to chemical weathering, result-ing in the formation of oxides, hydroxides, clay minerals andzeolites, depending on the geochemical environment.

Through the concentration values obtained for themajor cations(Ca, Mg, K and Na), the hydrogen ion (pH) and silicate (as colloidalH4SiO4) in the four sand pit lakes, in both seasons, it is noticed thatthe majority of the samples is in equilibriumwith kaolinite (Ca, Mgand Na diagrams), and only one sample from sand pit lake 1 has

y diagrams in the systems (CaO, MgO, K2O and Na2O)eAl2O3eSiO2eH2O at 27 �C and

Fig. 10. Sand pit lakes sampling plots (at the wet and the dry season) in the mineral stability diagrams in the K2OeMgOeAl2O3eSiO2-H2O system at 27 �C and 1 atm.

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188186

silicate values reaching the stability field of gibbsite during the wetseason. The stability fields of primary minerals are only pointed outin K diagrams, where muscovite takes place in the wet season andK-feldspar in the dry season. A possible explanation for this fact isthe weathering resistance of the K-minerals, also observed in the Kversus Mg diagram (Fig. 10), which shows with more details thepossible mineral assemblage in equilibrium with sand pit lakes. Ingeneral, it is observed that the contents of silicate and major ionshave slight variations between seasons, resulting in similar plotconfigurations for all diagrams (Fig. 9).

The features observed in the stability diagrams show thatweathering rating in these sand pit lakes may be classified asmoderated, that is, the chemical composition during the dry andthe wet seasons suggests equilibrium with kaolinite, besides theabnormal acidification.

6. Conclusions

The sand pit lakes form a peculiar environment due to theirwater acidification, originating an atypical water compositioncompared to natural water bodies and other mine pit lakes. Thepluvial regime of the region and the mining activities are theexternal controllers of physicalechemical parameters whichcontrol the dissolved species in pit lakes water. Concerning thephysicalechemical parameters, the EC could exert pH control,mainly in the dry season, when the evapoconcentration processtakes place due to high evapotranspiration, leading to high valuesof Ca, Mg, Na, SiO2 and Al in the water. The data presented in thiswork also show that the non-interruption of sand mining activities

makes the SO4 the best retriever of Al in the sand pit lakes in bothseasons, despite the important role of SiO2 during the dry season,precipitated as hydroxialuminosilicate.

The relatively high acidity and aluminum contents and the lownutrient concentrations suggest that these pit lakes waters shouldbe treated prior to use as fish ponds if sustainable economic reve-nues are expected. Therefore, any fish growing in these lakes willrequire pH correction, which will simultaneously reduce aluminumavailability. Fertilization, however, would also be required even forextensive farming practices, due to the oligotrophic nature of thewaters, in particular the very low phosphorus and nitrogencontents. The fish species to be introduced in the pit lakes can alsoinfluence the success of the aquaculture. Tilapia (Oreochromisniloticus), for example, will be reared without problems, and hasbeen used as a better alternative throughout the country (Conteet al., 2003).

A great concern is the possibility of Al groundwater contami-nation from sand pit lakes into surrounding aquifer. Values of pH inthe groundwater presented by Barbosa (2005) are higher comparedto sand pit lakes. These higher pH values together with largerresidence time of Al in aquifer porous framework precipitatealuminum as hydroxide (Al(OH)3 e hydrolysis reaction).

The lack of correlation between EC and pH together with theother aspects discussed, points out that the sand pit lakes behave asa system in relative hydrogeochemical equilibrium between waterand solid materials.

It is worth noting that the weathering assessment by mineralstability diagrams is valid only for the sand pit lakes, not for thewhole basin (aquifer), due to the abnormal acidification of those

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188 187

waters which accelerates sediment dissolution, masking the actualweathering rate of basin sediments.

Acknowledgements

This work is part of the Instituto Nacional de Ciência e Tecno-logia e INCT-TMCOcean (CNPq/573-601/2008-9). Eduardo D. Mar-ques would like to thank the National Research Council of Brazil(CNPq) and Fundação de Apoio a Pesquisa do Rio de Janeiro(FAPERJ) for his scholarship.

References

Alpers, C.N., Blowes, D.W., 1994. Environmental geochemistry of sulphide oxidation.In: American Chemical Society Symposium, Series 550. American ChemicalSociety, p. 681.

ASTM, 1992. Guide for conducting acute toxicity tests with fishes, macro-invertebrates and amphibians. In: Annual Book of ASTM Standards. StandardNo. E 729, vol. 11.04. American Society for Testing and Materials, Philapdelphia,pp. 383e384.

Bachmann, T.M., Friese, K., Zachmann, D.W., 2001. Redox and pH conditions in thewater column and in the sediments of an acidic mining lake. Journal ofGeochemical Exploration 73, 75e86.

Baird, C., 1998. Environmental Chemistry. W.H. Freeman and Company, New York,NY, USA, 557 pp.

Ball, J.W., Nordstrom, D.K., Zachmann, D.W., 1987. WATEQ4F e A Personal ComputerFORTRAN Translation of the Geochemical Model WATEQ2 with Revised DataBase. U.S. Geological Survey, Open-File Report 87-50, 108 pp.

Barbosa, C.F., 2005. Hidrogeoquímica e a contaminação por nitrato em água sub-terrânea no bairro Piranema, Seropédica e RJ. MSc Dissertation e (Geology andNatural Resources)-UNICAMP, Campinas, São Paulo, Brazil, 87 pp.

Bendell-Young, L.I., Harvey, H.H., 1992. The relative importance of manganese andiron oxides and organic matter in sorption of trace metals by surficial lakesediments. Geochimica et Cosmochimica Acta 56, 1175e1186.

Bendell-Young, L.I., Harvey, H.H., Dillon, P.J., Scholer, P.J., 1989. Contrasting behaviorof Mn in the surficial sediments of thirteen south-central Ontario lakes. Scienceof the Total Environment 87e88, 129e139.

Berbert, M.C., 2003. A mineração de areia no distrito areeiro de Itaguaí-Seropédica/RJ: geologia dos depósitos e caracterização das atividades de lavra e dosimpactos ambientais. MSc Dissertation, (Regional and Economic Geology), UFRJ,Rio de Janeiro, Brazil, 132 pp.

Bidone, E.D., Silva-Filho, E.V., Guerra, T., Barroso, L.V., Ovalle, A.R.C., 1999. Naturaland cultural levels in rivers of small coastal watersheds, S-SE e Brazil. In:Knoppers, B., Bidone, E.D., Abrão, J.J. (Eds.), Environmental Geochemistry ofCoastal Lagoon System, Rio de Janeiro, Brazil, first ed. EDUFF, Niterói, RJ, pp.89e105.

Birchall, J.D., Exley, C., Chappel, J.S., Phillips, M.J., 1989. Acute toxicity of aluminiumto fish eliminated in silicon-rich acid waters. Nature 338, 146e148.

Birge, W.J., 1978. Aquatic toxicology of trace elements of coal and fly ash. In:Thorp, J.H., Gibbons, J.W. (Eds.), Dep. Energy Symp. Ser.. Energy and Environ-mental Stress in Aquatic Systems, Augusta, GA, vol. 48, pp. 219e240.

Brockamp, O., 1976. Dissolution and transport of manganese by organic acids andtheir role in sedimentary Mn ore formation. Sedimentology 23, 579e586.

Camilleri, C., Markich, S.J., Noller, B.N., Turley, C.J., Parker, G., Van Dam, R.A., 2003.Silica reduces the toxicity of aluminium to a tropical freshwater fish (Mogurndamogurnda). Chemosphere 50, 355e364.

Campbell, P.G.C., Bisson, M., Bougie, R., Tessier, A., Villeneuve, J., 1983. Analyticalmethodology for determining the speciation of aluminum in acidic freshwaters.Analytical Chemistry 55 (14), 2246e2252.

Carvalho, D.F., Silva, L.D.B., Folegatti, M.V., Costa, J.R., Cruz, F.A., 2006. Evaluation ofthe reference evapotranspiration in Seropédica, Rio de Janeiro State, Brazil, usingweighing lysimeter. Revista Brasileira de Agrometeorologia 14 (2), 187e195.

Castendyk, D.N., Webster-Brown, J.G., 2007. Sensitivity analyses in pit lake predic-tion, Martha mine, New Zealand 2: geochemistry, watererock reactions, andsurface adsorption. Chemical Geology 244 (1e2), 56e73.

CCREM (Canadian Council of Ministers of the Environment), 1999. Canadian waterquality guidelines for the protection of agricultural water uses: introduction. In:Canadian Environmental Quality Guidelines, 1999. Canadian Council of Minis-ters of the Environment, Winnipeg.

Conte, L., Sonoda, D.Y., Shirota, R., Cyrino, J.E.P., 2003. Performance economics ofTilapia culture in Brazil. Global Aquaculture Advocate 6 (6), 57e58.

de Mello, W.Z., 2001. Precipitation chemistry in the coast of the MetropolitanRegion of Rio de Janeiro Brazil. Environmental Pollution 114, 235e242.

Denimal, S., Bertrand, C., Mudry, J., Paquette, Y., Hochart, M., Steinmann, M., 2005.Evolution of the aqueous geochemistry of mine pit lakes e Blanzy-Montceau-les-Mines coal basin (Massif Central, France): origin of sulfate contents; effectsof stratification on water quality. Applied Geochemistry 20, 825e839.

Deutsch, W.J., 1997. Groundwater Geochemistry e Fundamentals and Application toContamination. CRC Press LLC, Boca Raton, Florida, USA, 221 pp.

Drever, J.I., 1982. The Geochemistry of Natural Waters. Prentice-Hall, Inc., Engle-wood Cliffs, N.J., USA, 388 pp.

Driscoll, C.T., Schecher, W.D., 1990. The chemistry of aluminum in the environment.Environmental Geochemistry and Health 12 (1e2), 28e49.

Duan, J., Gregory, J., 1998. The influence of silicic acid on aluminium hydroxideprecipitation and flocculation by aluminium salts. Journal of InorganicBiochemistry 69, 193e201.

Eary, L.E., 1998. Predicting the effects of evapoconcentration on water quality inmine pit lakes. Journal of Geochemical Exploration 64, 223e236.

Eary, L.E., 1999. Geochemical and equilibrium trends in mine pit lakes. AppliedGeochemistry 14 (8), 963e987.

ELETROBOLT, 2003. Estudos Hidrogeológicos dos Aqüíferos Intergranulares a Oestedo Rio Guandu, Município de Seropédica/Rio de Janeiro, 275 pp.

Exley, C., Chappel, J.S., Birchall, J.D., 1991. A mechanism for acute aluminum toxicityin fish. Journal of Theoretical Biology 151, 417e428.

Fisher, T.S.R., Lawrence, G.A., 2000. Observations at the upper halocline of the IslandCopper pit lake. In: Larwrence, G.A., Pieters, R., Yonemistsu, N. (Eds.), Proc. ofthe 5th Int. Symp. Stratified Flows. Vancouver, BC, Canada, pp. 413e418.

Fleet, M.E., 1984. Short Course in Environmental Geochemistry. MineralogicalAssociation of Canada, London, 306 pp.

Gensemer, R.W., Playle, R.C., 1999. The bioavailability and toxicity of aluminum inaquatic environments. Critical Reviews in Environmental Science and Tech-nology 29, 315e450.

Góes, M.H.B., 1994. Sobre a Formação Piranema da Bacia do Rio Guandu (RJ): umestudo paleoambiental. In: 38 Congresso Brasileiro de Geologia, Camboriú,Brasil, pp. 322e323.

Grasshoff, K., Ehrhardt, M., Kremling, K., 1983. Methods of Seawater Analysis,second ed. Verlag Chemie, Weinheim, Germany, 419 pp.

Gray, N.F., 1996. Field assessment of acid mine drainage contamination in surfaceand ground water. Environmental Geology 27, 358e361.

Hem, J.D., 1985. Study and Interpretation of the Chemical Characteristics of NaturalWater, third ed.. US Geological Survey, Water Supply Paper 2254, 263 pp.

Jambor, J.L., Blowes, D., 1994. Environmental Geochemistry of Sulfidic Mine Wastes.Mineral Association of Canada, Nepean, Ontario, 438 pp.

Kennedy-Perkins, A., 2002. Rehabilitation at the Macraes gold project: from pit topark. In: Proc. of the Australasian Institute of Mining and Metallurgy AnnualConf. 2002, Publication Series No. 6/02, Auckland, New Zealand, pp. 41e43.

Klöppel, H., Fliedner, A., Kördel, W., 1997. Behaviour and ecotoxicology ofaluminium in soil and water e review of the scientific literature. Chemosphere35 (1e2), 353e363.

Larsen, D., Mann, R., 2005. Origin of high manganese concentrations in coal minedrainage, eastern Tennessee. Journal of Geochemical Exploration 86, 143e163.

Marques, E.D., Tubbs, D., Silva-Filho, E.V., October 25e29 2004. Hydrogeochemistrycharacterization of Piranema aquifer, Southeastern Brazil. In: 4th InternationalSymposium of Environmental Geochemistry in Tropical Countries, Búzios e Riode Janeiro, Brazil, pp. 119e120.

Marques, E.D., Sella, S.M., de Mello, W.Z., Lacerda, L.D., Silva-Filho, E.V., 2008.Hydrogeochemistry of Sand Pit Lakes at Sepetiba Basin, Rio de Janeiro, South-eastern Brazil. Water Air and Soil Pollution 189, 21e36.

McDonald, D.G., Reader, J.P., Dalziel, T.R.K., 1989. The combined effects of pH andtrace metals on fish ionoregulation. In: Morris, R., Taylor, E., Brown, D., Brown, J.(Eds.), Acid Toxicity and Aquatic Animals. Cambridge University Press, Boston,pp. 221e242.

Miretzky, P., Conzonno, V., Cirelli, A.F., 2001. Geochemical processes controllingsilica concentrations in groundwaters of the Salado River drainage basin,Argentina. Journal of Geochemical Exploration 73, 155e166.

Modis, K., Adam, K., Panagopoulos, K., Komtopoulos, A., 1998. Development andValidation of a geostatistical model for prediction of acid mine drainage inunderground sulphide mines. Transaction of the Institution of Mining andMetallurgy (Section A: Mining Industry) 107, 102e107.

Nordstrom, D.K., Ball, J.W., 1986. The geochemical behavior of aluminum in acidifiedsurface waters. Science 232, 54e56.

Ottoni, A.B., Silva, E.R., Giordano, G., Fortes, J.D.N., Pitta, V.L.J., 2002. A GestãoAmbiental Adequada da Bacia do Guandu. In: Seminário Bacia do Rio Guandu eProblemas e Soluções, Seropédica-RJ, Brazil, pp. 1e5.

Parkhurst, D.L., 1995. User’s Guide to PHREEQCeA Computer Program for Specia-tion, Reaction-path, Advective-transport, and Inverse Geochemical Calculations.U.S. Geological Survey, Water-Resources Investigations Report 95-4227, 143 pp.

Pellicori, D.A., Gammons, C.H., Poulson, S.R., 2005. Geochemistry and stable isotopecomposition of the Berkeley pit lake and surrounding mine waters, Butte,Montana. Applied Geochemistry 20 (11), 2116e2137.

Plumlee, G.S., Logsdon, M.J. (Eds.), 1998. The Environmental Geochemistry ofMineral Deposits. Reviews in Economic Geology, vol. 6A. Society of EconomicGeologists, Littleton, Colorado, pp. 224e240.

Poleo, A.B.S., 1995. Aluminium polymerisation e a mechanism of acute toxicity ofaqueous aluminium to fish. Aquatic Toxicology 31 (4), 347e356.

Ramstedt, M., Carlsson, E., Lövgren, L., 2003. The aqueous geochemistry in theUdden pit lake, northern Sweden. Applied Geochemistry 18, 97e108.

Rios, J.L.P., Berger, S.G., 2002. Estudos sócio-econômicos e de demanda de água paraa RMRJ. In: Seminário sobre a bacia hidrográfica do Rio Guandu: problemas esoluções, Seropédica, Rio de Janeiro, Brasil, pp. 26e32.

Robles-Arenas, V.M., Rodrigues, R., García, C., Manteca, J.I., Candela, L., 2006.Sulphide-mining impacts in the physical environment: Sierra de CartagenaeLaUnión (SE Spain) case study. Environmental Geology 51, 47e64.

SEMA, 1996. Diagnóstico ambiental da Bacia Hidrográfica da Baía de Sepetiba:Programa de Zoneamento Econômico-Ecológico do Estado do Rio de Janeiro.CARTOGEO/NCE/UFRJ, Rio de Janeiro, Brazil, 63 pp.

E.D. Marques et al. / Journal of South American Earth Sciences 30 (2010) 176e188188

Seinfeld, J.H., 1986. Atmospheric Chemistry and Physics of Air Pollution. John Wiley& Sons, New York, pp. 198e241.

Sheoran, A.S., Sheoran, V., 2006. Heavy metal removal mechanism of acid minedrainage in wetlands: a critical review. Minerals Engineering 19 (2), 105e116.

Silva-Filho, E.V., Barcellos, R.G.S., Emblanch, C., Blavoux, B., Sella, S.M., Daniel, M.,Simler, R., Wasserman, J.C., 2009. Groundwater chemical characterization ofa Rio de Janeiro coastal aquifer, SE e Brazil. Journal of South American EarthSciences 27, 100e108.

Stottmeister, U., Glässer, W., Klapper, H., Weißbrodt, E., Eccarius, B., Kennedy, C.,Schultze, M., Wendt-Potthoff, K., Frömmichen, R., Schreck, P., Strauch, G., 1999.Strategies for remediation of former opencast mining areas in eastern Germany.

In: Azcue, J.M. (Ed.), Environmental Impacts of Mining Activities. Springer,Berlin, pp. 263e296 (Chapter 16).

Stumm, W., Morgan, J.J., 1996. Aquatic Chemistry, Chemical Equilibria and Rates inNatural Waters, third ed. John Wiley & Sons, Inc., New York, NY, USA, 1022 pp.

Tubbs, D., 1999. Ocorrência das Águas Subterrâneas e “Aquífero Piranema” e

Município de Seropédica, área da Universidade Rural e Arredores, Estado do Riode Janeiro. FAPERJ (Fundação de Amparo a Pesquisa do Estado do Rio Janeiro),Brasil, Final Technical Report, 123 pp.

Ulrich, K.U., Rossberg, A., Foerstendorf, H., Zänker, H., Scheinost, A.C., 2006.Molecular characterization of uranium (VI) complexes on iron (III)-rich acidmine water colloids. Geochimica et Cosmochimica Acta 70 (22), 5469e5487.